This re-revised five-year proposal seeks funding for a twin-family study of adolescent alcohol use and abuse, and related behavioral risk-factors, using a prospective cohort-sequential design. Cohorts of female like-sex pairs, including minority twin pairs, who will be aged 13, 15, 17, and 19 by the time they are first studied, have been ascertained from state birth records (N=2350 twin pairs). Current addresses for the families will be traced, and zygosity information obtained by abbreviated telephone interview or mailed questionnaire (N= 1785 pairs). Over a four-year period, telephone screening interviews will be conducted with parents of the twins (N = 1500 families) and with twin pairs aged 17 and older (N=8O0 pairs). Approximately 310 pairs and their parent(s) will be randomly assigned for in-person interview, and a finder 190 'high-risk' families will be assigned for interview on the basis of parental or twin history of alcohol abuse or dependence. Annual abbreviated follow-up telephone interviews will be conducted with all available twin pairs at ages 16-22, and with the parents of twins aged 15; and biannual in-person follow-up interviews of the parents and twins from the 'high-risk' and random control families will also be conducted. We will assess in both the twins and their parents multiple measures of alcohol use (quantity/frequency/ abstinence, frequency of drinking to intoxication, maximum consumption, drinking-related psychosocial and other problems, abuse, dependence); associated behavioral and psychiatric risk-factors (conduct problems, conduct disorder, oppositionaI disorder, ADHD, depression, anxiety disorders, history of other drug use, academic failure); and postulated mediating and/or moderating variables (positive and negative alcohol expectancies; values; measures of temperament and personality; family environment; perceived environment). Methods of multivariate genetic analysis will be used to test hypotheses concerning the developmental pathways by which genetic risk for alcoholism and environmental risk- factors operate; to identify critical mediating and moderator variables in these developmental pathways; and to explore the joint and correlated action ('genotype-environment correlation') and interaction ('genotype x environment interaction') of genetic and environmental risk-factors. The project will thus address a critical gap in our knowledge of the etiology of adolescent drinking behavior.